Method for monitoring network distributed computing resources using distributed cellular agents

• US 6,308,208 B1
• Filed: 09/30/1998
• Issued: 10/23/2001
• Est. Priority Date: 09/30/1998
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: McNally, Michael, Jung, Carey L.
• Primary Examiner(s): Geckil, Mehmet B.
• Application Number:
  US09/164,019
• Time in Patent Office: 1,119 Days
• Field of Search: 709/201, 709/202, 709/205, 709/223, 709/224, 709/204, 345/330, 370/254, 710/17, 710/18, 714/39, 714/47, 714/51
• US Class Current: 709/224
• CPC Class Codes: G06F 11/3006 where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems multiprogramming arrangements G06F9/46; allocation of resources G06F9/50
• CPC Class Codes: G06F 11/3055 Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available error or fault processing without redundancy G06F11/0703; error detection or correction by redundancy in data representation G06F11/08; error detection or correction by redundancy in operation G06F11/14; error detection or correction by redundancy in hardware G06F11/16
• CPC Class Codes: G06F 11/3079 the data filtering being achieved by reporting only the changes of the monitored data
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: H04L 41/044 comprising hierarchical management structures
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/12 Network monitoring probes

Abstract

A method of automated distributed resource monitoring in a large distributed computing environment, wherein a given master resource comprises a set of given computing resources. The method begins by associating a set of one or more “cells” with a set of given computing resources that comprise the master resource. Each cell preferably is associated with a respective one of the set of given computing resources and has a set of one or more attributes whose values collectively define a “state” of the cell. Whenever a change in an attribute of a given cell effects a change in that cell'"'"'s state, the attribute change is propagated across each cell directly impacted by the cell state change, as well as to those observing cells that may be indirectly affected.

Claims

1. A method of resource monitoring in a large distributed computing environment, wherein a given master resource comprises a set of given computing resources, comprising the steps of:

(a) associating a set of cells with a set of given computing resources that comprise the master resource, each cell associated with a respective one of the set of given computing resources and having a set of one or more attributes whose values collectively define a state of the cell;

(b) upon change of a given attribute of a given cell that effects a change in cell state, propagating the attribute change across each cell directly impacted by the cell state change; and

(c) propagating resulting cell state changes across any affected cells.

2. The method as described in claim 1 wherein each cell uniquely identifies the given computing resource.

3. The method as described in claim 1 wherein a given cell attribute value that is a function of a constant.

4. The method as described in claim 1 wherein a given cell attribute value is a function of a primitive operation.

5. The method as described in claim 1 wherein a given cell attribute value is a function of other cell attributes.

6. The method as described in claim 1 wherein a given cell attribute value is a function of attributes of one or more other cells.

7. The method as described in claim 1 wherein a cell is a software agent deployed to the given computing resource.

8. The method as described in claim 7 wherein the software agent is executed in a runtime engine.

9. A method of resource monitoring in a distributed computer network having a management server servicing a set of machines, comprising the steps of:

deploying instances of a runtime engine across a subset of the machines to create a distributed runtime environment in the distributed computer network;

associating a set of one or more agents with the subset of the machines, each agent associated with a respective one of the set of given computing resources and having a set of one or more attributes whose values collectively define a state of the agent; and

upon change of a given attribute of a given agent that effects a change in agent state, propagating the attribute change across each agent directly or indirectly impacted by the agent state change.

10. The method as described in claim 9 wherein a given agent attribute value is a function of a constant.

11. The method as described in claim 9 wherein a given agent attribute value is a function of a primitive operation.

12. The method as described in claim 9 wherein a given agent attribute value is a function of other attributes of that agent.

13. The method as described in claim 9 wherein a given agent attribute value is a function of attributes of one or more other agents.

14. A method of automated distributed monitoring in a large distributed computing environment, comprising the steps of:

(a) associating a given computing resource in the distributed computing environment with a monitoring cell, wherein each monitoring cell has a set of one or more attributes whose values collectively define a state of that cell;

(b) organizing the set of monitoring cells into a hierarchy having at least a first level and a second level, wherein at least one cell in the first level has attributes that depend on attributes in one or more cells in the second level; and

(c) upon a change of state of a given cell in the second level, automatically propagating new attribute values to at least one cell in the first level.

15. The method as described in claim 14 further including updating the state of the cell in the first level to reflect the new attribute values propagated from the cell in the second level.

16. A system for use in a distributed computer network having a management server servicing a set of managed computers, comprising:

a plurality of software agents for implementing a resource monitoring task, each software agent deployed at a given computing resource and having a set of one or more attributes whose values collectively define a state of that software agent;

a plurality of instances of a runtime environment installed at given managed computers, wherein the runtime environment includes a runtime engine for executing a software agent to effect the resource monitoring task; and

means associated with a given software agent and responsive to a change of state of that software agent for automatically propagating new attribute values to at least one other software agent.

17. The system as described in claim 16 wherein the runtime engine is a virtual machine in a browser.

18. The system as described in claim 17 wherein the software agent is an applet.

19. A computer program product for use in a computer connected within a distributed computing environment having a management server servicing a set of managed computers, comprising:

instruction means for implementing a resource monitoring task as a function of a set of one or more attributes whose values collectively define a state;

control means responsive to a change of the state for automatically propagating new attribute values to at least one other computer implementing the resource monitoring task.

20. A computer connected within a distributed computing environment having a management server servicing a set of managed computers, comprising:

a processor;

an operating system;

instruction means executed by the operating system for implementing a resource monitoring task as a function of a set of one or more attributes whose values collectively define a state; and

control means responsive to a change of the state for automatically propagating new attribute values to at least one other computer within the distributed computing environment implementing the resource monitoring task.